Due to improvements in obstetrical and neonatal care, and an associated decrease in perinatal mortality, as well as increased risks associated with in vitro fertilization procedures, the number of preterm (< 37 weeks postmenstrual age) very low birth weight (< 1,500 g) neonates born is growing [34]. Of the approximately 50,000 neonates born yearly in the United States with a low birth weight, approximately 85% survive. These infants have an increased risk of long-term developmental disabilities, with 5-15% exhibiting major neurodevelopmental disabilities (e.g., cerebral palsy, mental retardation) and an additional 25-50% exhibiting mild to moderate neurodevelopmental disabilities (e.g., school failure) [40, 74]. This SBIR Phase II grant application is concerned with the detection of developmental disorders at a very early age. Assessment of the consequences of low birth weight, adverse pregnancies, or complicated births on the structure and function of an infant's central nervous system (CNS) and its development is important for obtaining valid estimates of behavior and neurological status, documenting recovery from acute eects, predicting subsequent outcome, and suggesting specic intervention strategies for any problems noted. Although brain imaging, such as cranial ultrasound, computed tomography, and magnetic resonance imaging, can document structural brain damage, assessment of infants' endogenously-generated (i.e., spontaneous) movement patterns, known as \general movements (GMs), provides a method by which to assess CNS function. In the past 25 years, an extensive body of literature has shown that motility, rst of the fetus and subsequently of the neonate, has strong clinical signicance as an indicator of brain function and dysfunction. Presently, assessment of GMs is based on the Gestalt evaluation of video-recorded infant movement patterns. Assessment of GMs is a useful adjunct to the standard neurological examination, primitive reex prole, and assessment of muscle strength and tone. However, the use of video recording is not ideal; it is tedious, subjective, labor intensive, and expensive, requiring specially-trained practitioners to evaluate stored image sequences o line. The proposed GEneral Movements for the Identication of Neurological disorders in Infants (GEMINI) sys- tem, a low-cost tool for objectively and automatically assessing infant GMs in hospital and non-hospital settings, will overcome these limitations. The GEMINI system will exploit Barron Associates' patented six degrees-of- freedom motion monitoring capability and sophisticated signal processing to identify the key qualitative and quantitative attributes of GMs, and thereby automate the assessment of GMs. Based on the proposed multi- phase research program, the requisite knowledge will be obtained to enable realization of a low-cost, low-prole GEMINI product that can reliably perform the required discrimination and serve as a valuable research tool in documenting and better understanding movements and movement precursors for neurological disorders.